Total runs:
326.5K
Run Growth:
-763.2K
Growth Rate:
-233.73%
Sponsored by
Tripo AI
tiiuae / Falcon3-3B-Instruct
Model's Last Updated: January 10 2025
text-generation
Introduction of Falcon3-3B-Instruct
Model Details of Falcon3-3B-Instruct
Falcon3-3B-Instruct
Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B parameters.
Falcon3-3B-Instruct achieves strong results on reasoning, language understanding, instruction following, code and mathematics tasks. Falcon3-3B-Instruct supports 4 languages (English, French, Spanish, Portuguese) and a context length of up to 32K.
Model Details
-
Architecture
- Transformer-based causal decoder-only architecture
- 22 decoder blocks
- Grouped Query Attention (GQA) for faster inference: 12 query heads and 4 key-value heads
- Wider head dimension: 256
- High RoPE value to support long context understanding: 1000042
- Uses SwiGLU and RMSNorm
- 32K context length
- 131K vocab size
- Pruned and healed from Falcon3-7B-Base on only 100 Gigatokens of datasets comprising of web, code, STEM, high quality and mutlilingual data using 1024 H100 GPU chips
- Posttrained on 1.2 million samples of STEM, conversational, code, safety and function call data
- Supports EN, FR, ES, PT
- Developed by Technology Innovation Institute
- License: TII Falcon-LLM License 2.0
- Model Release Date: December 2024
Getting started
Click to expand
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "tiiuae/Falcon3-3B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "How many hours in one day?"
messages = [
{"role": "system", "content": "You are a helpful friendly assistant Falcon3 from TII, try to follow instructions as much as possible."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=1024
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)
Benchmarks
We report in the following table our internal pipeline benchmarks.
- We use lm-evaluation harness .
- We report raw scores obtained by applying chat template without fewshot_as_multiturn (unlike Llama3.1).
- We use same batch-size across all models.
| Category | Benchmark | Llama-3.2-3B-Instruct | Qwen2.5-3B-Instruct | Nemotron-Mini-4B-Instruct | Falcon3-3B-Instruct |
|---|---|---|---|---|---|
| General | MMLU (5-shot) | 29.3 | 56.2 | 56.4 | 55.7 |
| MMLU-PRO (5-shot) | 11.9 | 17.2 | 23.3 | 29.7 | |
| IFEval | 73.9 | 64.2 | 66.5 | 68.3 | |
| Math | GSM8K (5-shot) | 68.5 | 58.5 | 46.9 | 71.9 |
| GSM8K (8-shot, COT) | 74.5 | 64.0 | 46.5 | 71.6 | |
| MATH Lvl-5 (4-shot) | 2.4 | 0.0 | 0.0 | 19.9 | |
| Reasoning | Arc Challenge (25-shot) | 38.9 | 50.0 | 51.2 | 58.5 |
| GPQA (0-shot) | 28.1 | 29.2 | 27.0 | 29.6 | |
| GPQA (0-shot, COT) | 11.3 | 11.0 | 12.2 | 26.5 | |
| MUSR (0-shot) | 34.9 | 40.2 | 38.9 | 39.0 | |
| BBH (3-shot) | 33.1 | 44.1 | 38.1 | 45.4 | |
| CommonSense Understanding | PIQA (0-shot) | 74.6 | 73.8 | 74.6 | 75.6 |
| SciQ (0-shot) | 77.2 | 60.7 | 71.0 | 95.5 | |
| Winogrande (0-shot) | - | - | - | 65.0 | |
| OpenbookQA (0-shot) | 40.8 | 41.2 | 43.2 | 42.2 | |
| Instructions following | MT-Bench (avg) | 7.1 | 8.0 | 6.7 | 7.2 |
| Alpaca (WC) | 19.4 | 19.4 | 9.6 | 15.5 | |
| Tool use | BFCL AST (avg) | 85.2 | 84.8 | 59.8 | 65.3 |
| Code | EvalPlus (0-shot) (avg) | 55.2 | 69.4 | 40.0 | 52.9 |
| Multipl-E (0-shot) (avg) | 31.6 | 29.2 | 19.6 | 32.9 |
Useful links
- View our release blogpost .
- Feel free to join our discord server if you have any questions or to interact with our researchers and developers.
Technical Report
Coming soon....
Citation
If the Falcon3 family of models were helpful to your work, feel free to give us a cite.
@misc{Falcon3,
title = {The Falcon 3 Family of Open Models},
url = {https://huggingface.co/blog/falcon3},
author = {Falcon-LLM Team},
month = {December},
year = {2024}
}
Runs of tiiuae Falcon3-3B-Instruct on huggingface.co
6.5K
Total runs
0
24-hour runs
-83
3-day runs
-680
7-day runs
-819
30-day runs
More Information About Falcon3-3B-Instruct huggingface.co Model
Url of Falcon3-3B-Instruct
Falcon3-3B-Instruct huggingface.co Url
Provider of Falcon3-3B-Instruct huggingface.co
Other API from tiiuae
Updated:October 12 2024
Total runs:
105.9K
Run Growth:
11.1K
Growth Rate:
10.49%
Total runs:
60.1K
Run Growth:
32.3K
Growth Rate:
53.81%
Total runs:
44.6K
Run Growth:
-125.2K
Growth Rate:
-280.36%
Total runs:
31.4K
Run Growth:
2.8K
Growth Rate:
8.79%
Total runs:
21.2K
Run Growth:
-2.7K
Growth Rate:
-12.84%
Total runs:
16.6K
Run Growth:
-508
Growth Rate:
-3.06%
Total runs:
15.8K
Run Growth:
-1.9K
Growth Rate:
-12.24%
Total runs:
15.3K
Run Growth:
1.1K
Growth Rate:
6.85%
Total runs:
14.7K
Run Growth:
-9.1K
Growth Rate:
-61.46%
Total runs:
13.3K
Run Growth:
-3.5K
Growth Rate:
-26.36%
Total runs:
11.7K
Run Growth:
-554
Growth Rate:
-4.74%
Total runs:
8.6K
Run Growth:
5.6K
Growth Rate:
64.38%
Total runs:
8.4K
Run Growth:
1.6K
Growth Rate:
18.87%
Total runs:
8.0K
Run Growth:
341
Growth Rate:
4.24%
Total runs:
7.5K
Run Growth:
1.5K
Growth Rate:
20.17%
Total runs:
5.0K
Run Growth:
1.9K
Growth Rate:
37.53%
Total runs:
4.6K
Run Growth:
2.1K
Growth Rate:
45.56%
Total runs:
4.3K
Run Growth:
2.7K
Growth Rate:
62.75%
Total runs:
4.0K
Run Growth:
1.4K
Growth Rate:
34.67%
Total runs:
3.7K
Run Growth:
-967
Growth Rate:
-25.92%
Total runs:
3.2K
Run Growth:
31
Growth Rate:
0.96%
Total runs:
3.1K
Run Growth:
1.7K
Growth Rate:
56.84%
Total runs:
2.6K
Run Growth:
1.7K
Growth Rate:
66.82%
Total runs:
2.3K
Run Growth:
1.8K
Growth Rate:
80.17%
Total runs:
1.7K
Run Growth:
132
Growth Rate:
7.99%
Total runs:
1.6K
Run Growth:
636
Growth Rate:
40.80%
Total runs:
1.3K
Run Growth:
-2
Growth Rate:
-0.13%
Total runs:
1.3K
Run Growth:
851
Growth Rate:
65.61%
Total runs:
1.3K
Run Growth:
18
Growth Rate:
1.40%
Total runs:
1.3K
Run Growth:
378
Growth Rate:
29.42%
Total runs:
1.3K
Run Growth:
952
Growth Rate:
76.16%
Total runs:
988
Run Growth:
0
Growth Rate:
0.00%
Total runs:
986
Run Growth:
36
Growth Rate:
3.65%
Total runs:
914
Run Growth:
-396
Growth Rate:
-43.33%
Total runs:
851
Run Growth:
106
Growth Rate:
12.46%
Total runs:
769
Run Growth:
-363
Growth Rate:
-47.20%
Total runs:
683
Run Growth:
-200
Growth Rate:
-29.28%
Total runs:
532
Run Growth:
18
Growth Rate:
3.38%
Total runs:
525
Run Growth:
522
Growth Rate:
99.43%
Total runs:
500
Run Growth:
-33
Growth Rate:
-6.60%
Total runs:
482
Run Growth:
-23
Growth Rate:
-4.77%
Total runs:
473
Run Growth:
81
Growth Rate:
17.12%
Total runs:
446
Run Growth:
20
Growth Rate:
4.48%
Total runs:
400
Run Growth:
109
Growth Rate:
27.25%
Total runs:
385
Run Growth:
131
Growth Rate:
34.03%
Total runs:
383
Run Growth:
-51
Growth Rate:
-13.32%
Total runs:
327
Run Growth:
20
Growth Rate:
6.12%
Total runs:
324
Run Growth:
-231
Growth Rate:
-71.30%
Total runs:
317
Run Growth:
4
Growth Rate:
1.26%
Total runs:
311
Run Growth:
17
Growth Rate:
5.47%
Total runs:
306
Run Growth:
274
Growth Rate:
89.54%
Total runs:
293
Run Growth:
45
Growth Rate:
15.36%
Total runs:
273
Run Growth:
246
Growth Rate:
90.11%
Total runs:
223
Run Growth:
-212
Growth Rate:
-95.07%
Total runs:
201
Run Growth:
82
Growth Rate:
40.80%
Total runs:
200
Run Growth:
-433
Growth Rate:
-216.50%
Total runs:
193
Run Growth:
-24
Growth Rate:
-12.44%
Total runs:
184
Run Growth:
84
Growth Rate:
49.12%
Total runs:
181
Run Growth:
-80
Growth Rate:
-44.20%
Total runs:
176
Run Growth:
74
Growth Rate:
42.05%
Total runs:
175
Run Growth:
47
Growth Rate:
26.86%
Total runs:
167
Run Growth:
7
Growth Rate:
4.19%
Total runs:
166
Run Growth:
12
Growth Rate:
7.23%
Total runs:
154
Run Growth:
64
Growth Rate:
41.56%
Total runs:
147
Run Growth:
60
Growth Rate:
40.82%
Total runs:
136
Run Growth:
15
Growth Rate:
11.03%
Total runs:
107
Run Growth:
-231
Growth Rate:
-215.89%
Total runs:
105
Run Growth:
19
Growth Rate:
18.10%
Total runs:
105
Run Growth:
25
Growth Rate:
23.81%
Total runs:
97
Run Growth:
97
Growth Rate:
100.00%
Total runs:
77
Run Growth:
24
Growth Rate:
31.17%
Total runs:
72
Run Growth:
72
Growth Rate:
100.00%
Total runs:
62
Run Growth:
7
Growth Rate:
11.29%
Total runs:
60
Run Growth:
-4
Growth Rate:
-6.67%
Total runs:
59
Run Growth:
-149
Growth Rate:
-252.54%
Total runs:
55
Run Growth:
21
Growth Rate:
38.18%
Total runs:
52
Run Growth:
0
Growth Rate:
0.00%
Total runs:
50
Run Growth:
1
Growth Rate:
2.00%
Total runs:
45
Run Growth:
-55
Growth Rate:
-122.22%
Total runs:
42
Run Growth:
16
Growth Rate:
38.10%
Total runs:
40
Run Growth:
15
Growth Rate:
37.50%
Total runs:
40
Run Growth:
17
Growth Rate:
42.50%
Total runs:
39
Run Growth:
9
Growth Rate:
23.08%